Multi-drug resistant Gram negative bacteria (MDR-GNB) present a significant threat to successful antimicrobial chemotherapy. Among MDR-GNB, extended spectrum beta-lactamase (ESBL) producing Klebsiella pneumoniae resistant to ceftazidime are associated with increased mortality. The atomic structure determination of the SHV-2 beta-lactamase, the first ESBL described, revealed the active site of the enzyme is widened to accommodate the side chain of ceftazidime. Despite this insight, our understanding is still incomplete. The broad, long-term objectives of this proposal are: 1) To determine the atomic structures of boronic acid transition-state analogue inhibitors that have the R1 side chains of ceftazidime and cefotaxime in the SHV-1, -2 and -5 beta-lactamases. This will elucidate the critical interactions that define the ability of ESBLs to hydrolyze advanced generation cephalosporins. 2) Using strains of E. coli deficient in the principal DNA mismatch repair protein, mutS, select blaSHV mutants that stabilize and enhance expression of SHV beta-lactamase. 3) To determine if mutator phenotypes exist in K. pneumoniae;assess the relationship between ESBL enzymes, mutator phenotypes, and mutS expression. 4) To develop a model for ESBL evolution using blaSHV in a mutS deficient strain of E. coli. Accomplishing these goals will achieve an unprecedented understanding of the protein and substrate interactions and genetic properties responsible for the ESBL phenotype. This work will also serve as a paradigm for the prediction of novel ESBL phenotypes in SHV enzymes, form a basis for the evaluation of novel beta-lactams, and show how enzyme drug targets influence substrate affinity and catalysis.